meg51/whisper-large-v3-hindi

Whisper Large-v3 Hindi -megha sharma

This model is a fine-tuned version of openai/whisper-large-v3 on the Google Fleurs dataset. It achieves the following results on the evaluation set:

• Loss: 0.1607
• Wer: 18.4303

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-06
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2000
• training_steps: 20000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1781	6.7797	2000	0.1785	21.1734
0.1519	13.5593	4000	0.1621	19.2405
0.1286	20.3390	6000	0.1577	18.7427
0.1259	27.1186	8000	0.1564	18.2058
0.111	33.8983	10000	0.1568	17.9032
0.1067	40.6780	12000	0.1582	17.8153
0.1034	47.4576	14000	0.1591	18.8403
0.0995	54.2373	16000	0.1603	18.8598
0.0929	61.0169	18000	0.1607	18.4303

Framework versions

• PEFT 0.12.1.dev0
• Transformers 4.45.0.dev0
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1